Online dispute resolution for crypto-asset disputes: AI, smart contracts and digital inclusion by design

Crypto-asset disputes create distinctive barriers to effective remedies because they combine technical opacity, cross-border legal fragmentation, pseudonymous actors, irreversible transactions and uncertain enforcement pathways. These barriers are especially significant for vulnerable and non-expert users, who may be unable to identify the legally relevant harm, organize blockchain-based evidence, select an appropriate forum or obtain practical enforcement. This article develops a conceptual, normative and legal-design proposal for a hybrid Online Dispute Resolution (ODR) architecture that integrates off-chain artificial intelligence (AI) assistance with a minimized on-chain smart-contract layer. The model assigns AI systems a supporting role in legal information, triage, claim classification, evidence organization, explanation and negotiation, while reserving contested legal determinations for human review. A permissioned distributed ledger records only hashes, timestamps, procedural events, escrow states and execution proofs; personal data, communications, evidence and AI processing remain off-chain. Smart contracts are used as bounded enforcement tools for registration, escrow and execution of validated settlement terms, rather than as autonomous adjudicators. The article applies the architecture to three illustrative scenarios: asset seizure and recovery, exchange failure and mass claims, and crypto scams involving cross-border fraud. It then evaluates the model against European requirements concerning artificial intelligence, crypto-asset services, data protection, procedural fairness, cybersecurity and accountability. The contribution is twofold. First, it advances the concept of access to remedies by design, under which comprehension, participation, contestability, evidentiary integrity and enforceability are embedded into the architecture of digital dispute resolution. Second, it provides a technically specified evaluation framework for future prototyping and empirical validation. The article does not claim measured improvements in accessibility or efficiency; rather, it identifies testable design requirements and performance indicators for subsequent pilots, user studies and regulatory sandboxes.